Restoring finger dexterity with an exoskeleton controlled by human intracranial recordings

In Brain-Computer Interfacing (BCI), brain activity is recorded and translated into actions intended by the user. Recent developments capitalize on the relation between motor actions and localized activity of motor- and somatosensory regions of the brain. The ability to control a robotic limb or regain control over a paralyzed limb with a motor-BCI has promoted the latter as a solution for patients devoid of voluntary movement control. Electrocorticography (ECoG), a partially invasive recording technique, offers new perspectives for such challenging applications as it avoids scarring or other histological processes and combines high spatio-temporal resolution and broad bandwidth with long-term recording stability. What is still lacking is the self-paced control of individual fingers from imagined motor activity, crucial to serve the targeted patients. This also summarizes the first objective of this project: to develop a robust and accurate model for decoding intended finger trajectories from ECoG recordings. Nevertheless, motor imagery is a skill that needs to be learnt, hence, our second objective: to develop a multi-step training strategy that will assist the user in gradually acquiring imagined finger control. We will focus on a hand exoskeleton, a wearable motorized framework that supports finger movements, from which we expect beneficial training effects as it the provides the user with tactile and proprioceptive feedback of the intended actions.